Mapping dimensions of biodiversity in forested ecosystems with lidar and imaging spectroscopy fusion

Abstract:

To advance science and policy in the face of rapid environmental change, researchers and practitioners need access to consistent monitoring information about how ecosystems are changing. Remote sensing tools have the potential to facilitate mapping and understanding of biodiversity variation across spatiotemporal scales. Despite this, inconsistencies in the characterization and definitions of ecosystem heterogeneity exist among the biodiversity and remote sensing communities, leading to large uncertainties in the relationships between remotely sensed metrics and biodiversity measurements. Here we use imaging spectroscopy and lidar data from the National Ecological Observatory Network’s Airborne Observation Platform (NEON AOP), as well as NEON field observations of tree diversity, to test spectral and structural diversity metrics across a range of forest types. We find that, within the US eastern temperate forest biome, structural metrics better predict taxonomic and functional biodiversity, but phylogenetic diversity is more closely tied to spectral diversity and inter-site differences. With individual plots as the unit of analysis, tree diversity is not closely tied to latitude, and measures of geodiversity are also important predictors of tree diversity. This work highlights the importance of considering spatial scale and biogeography in determining connections between remote sensing and different facets of biodiversity.